Testing, Testing: A/B Testing For Referral Programs

von Sean Ogino |

Testing, Testing: A/B Testing For Referral Programs

Whether it’s a life or business, it rarely happens that you begin something, let’s say a program or application with a new approach, and it starts to work well from the moment it hits the point of implementation. You have to think about different options that might help your cause. Then you constantly observes it, compare it with other results, analyze those results and then you include the best elements in your application or approach. Referral marketing is no different. It also has to go through the process of molding and remolding before it takes a final and concrete form. And for this purpose, marketers use AB (alternatively written as A/B) tests to ensure that the program has been built by all well performing and well-nourished building blocks.

AB testing, also known as split testing, is a comparison between two versions of a webpage or application or program against each other in order to determine which one performs better. It is decided by various qualitative and quantitative metrics with the sole goal of bottom line improvement. Thus, they are extremely important. Also of importance in e-commerce are sharing and referral solutions–they’re the most trusted form of advertising and give you customers that have 25% higher lifetime values. Consequently, we’ll talk today about the vital variables for AB testing for referral programs.

A) Referral Call-to-Action Location on Home Page: The call to action is exceptionally valuable as this is the first step that customers perform when they are convinced by your offerings. The same applies to referral CTAs too. Hence, testing everything about it, like location and color, makes a lot of sense. You can have two versions as far as the location of the CTA button is concerned. You can link it to your referral program with a button on your upper navigation bar, just like Try the World does.

try the world top nav raf slide

You can also experiment with putting the CTA at the bottom of the page, or in the My Account dropdown, like Vitacost does.

Vitacost raf account promotion slide

All you have to do is to compare the results of both the version and stick with the version that is yielding better results. You can do another test by trying different colors of the CTA buttons.

B) Reward Size and Type: It’s a myth that bigger rewards always drive more people to become your advocates. There are plenty of examples where smaller rewards have proved more effective than the bigger ones. The truth is, many people feel that they are helping their friends by suggesting products. It’s not necessary that all of your customers will look at your referral program as a ploy to get some rewards. The same is true with the reward types. We have seen that a reward that has worked for a few companies hasn’t performed as well for other companies despite being in the same segment. There is nothing like a thumb rule in this. That’s why the only way to know what will work for your unique audience is to perform AB testing for Empfehlungsprogramme incentives.

C) Landing Page Copy: You can test the copy on your referral landing page too. The way you express yourself in your referral programs can have different effects on different people. That’s precisely why testing the landing page copy can help you. Airbnb did this copy test with reference to its email copy that it sends to its customers promoting referral. As we all know, Airbnb’s referral program is double sided in which both advocates and their friends receive $25 in credit. Half of its customers got an email accentuating their reward, and the other half got one focusing the friend’s reward. It found that the most unselfish option was the clear winner.

Refer a friend

D) Personalization: Though the awareness about behavioral targeting is getting more traction nowadays, it is still underused by many referral marketers. And that’s why its use can assure you the competitive advantage. In simple terms, personalization means communicating with your customers in a manner in which they are comfortable. It can be as simple as targeting their locations or their current stage in life.

For example, if you know that your customers spend more time on mobile, you can send all the important updates or promotional emails to their mobiles. Or, if you know that your target customer is college going crowd, you can create referral programs which are centralized around students. You can come up with a referral program that offers special treatment or rewards to students. Besides, you can also test at what time people are more prone to respond to the referral sent by your customers. You can test aforementioned and various other methods of personalization and see which one works for you.

E) Ways To Refer A Friend: The entire point of having referral program is to motivate more people to refer your products to their friends. There are multiple ways through which your customers can do that- email, social media, and receipt codes are just a few of the options. As every business is different and every company’s target customers are different, you should check which mode is giving you more satisfactory results. Johnnie O, for example, lets users share via social media, email, link copying and pasting, or code copying and pasting.

johnnie o raf sharing options

F) Videos: It may look strange, but considering the way in which videos are taking over every available digital space, it’s not a wrong idea to test how you can use videos for the promotion of your referral program. It may be in the form of advertisement or explaining the benefits and the overall functioning of your referral program via videos. You can A/B test your actual videos to see which performs best and leads to more referrals. The untested videos will continue to exist when there is a total lack of any sort of robust testing.  It can easily drive potential customers away from you and negatively affect your referral promotion efforts. So, create videos, test them and go on with the best working ones.

G) Adding Languages: Of course, this might not be a regular member in your list of AB testing for referral programs priorities, but there are instances where the multilingual approach has worked. There are places in the USA where the Chinese and Latin population is significant, for example. If you want to tap that customer zone, you can try to communicate with them in their own languages as an addition to your English referral program. If the outcome is positive, you can seriously stick with it.

Check out this Dutch example from Vivobarefoot, who also has versions of their referral program in English, French, Spanish, and German.

vivo international implementations slide

Zum Schluss... A/B testing is brutal and in-depth scrutiny of your plan and program. It is nothing but a process to ensure that you have checked all the available paths and then deciding which one is yours. And as they say, the first prerequisite of reaching your decided destination is to know the path unequivocally. That’s why A/B testing is of enormous importance!

[White Paper]
[White Paper]